arXiv ·
Researchers from MBZUAI have released MobiLlama, a fully transparent open-source 0.5 billion parameter Small Language Model (SLM). MobiLlama is designed for resource-constrained devices, emphasizing enhanced performance with reduced resource demands. The full training data pipeline, code, model weights, and checkpoints are available on Github.
arXiv ·
The paper introduces Sparse-Quantized Representation (SpQR), a new compression format and quantization technique for large language models (LLMs). SpQR identifies outlier weights and stores them in higher precision while compressing the remaining weights to 3-4 bits. The method achieves less than 1% accuracy loss in perplexity for LLaMA and Falcon LLMs and enables a 33B parameter LLM to run on a single 24GB consumer GPU. Why it matters: This enables near-lossless compression of LLMs, making powerful models accessible on resource-constrained devices and accelerating inference without significant accuracy degradation.
arXiv ·
This paper introduces SimulMask, a new paradigm for fine-tuning large language models (LLMs) for simultaneous translation. SimulMask utilizes a novel attention masking approach that models simultaneous translation during fine-tuning by masking attention for a desired decision policy. Applied to a Falcon LLM on the IWSLT 2017 dataset, SimulMask achieves improved translation quality compared to state-of-the-art prompting optimization strategies across five language pairs while reducing computational cost. Why it matters: The proposed method offers a more efficient way to adapt LLMs for real-time translation, potentially enhancing multilingual communication tools and services.
arXiv ·
This paper introduces a minimalistic autonomous racing stack designed for high-speed time-trial racing, emphasizing rapid deployment and efficient system integration with minimal on-track testing. Validated on real speedways, the stack achieved a top speed of 206 km/h within just 11 hours of practice, covering 325 km. The system performance analysis includes tracking accuracy, vehicle dynamics, and safety considerations. Why it matters: This research offers insights for teams aiming to quickly develop and deploy autonomous racing stacks with limited track access, potentially accelerating innovation in autonomous vehicle technology within the A2RL and similar racing initiatives.
MBZUAI ·
A presentation discusses using programmable network devices to reduce communication bottlenecks in distributed deep learning. It explores in-network aggregation and data processing to lower memory needs and increase bandwidth usage. The talk also covers gradient compression and the potential role of programmable NICs. Why it matters: Optimizing distributed deep learning infrastructure is critical for scaling AI model training in resource-constrained environments.
arXiv ·
The paper introduces LLMEffiChecker, a tool to test the computational efficiency robustness of LLMs by identifying vulnerabilities that can significantly degrade performance. LLMEffiChecker uses both white-box (gradient-guided perturbation) and black-box (causal inference-based perturbation) methods to delay the generation of the end-of-sequence token. Experiments on nine public LLMs demonstrate that LLMEffiChecker can substantially increase response latency and energy consumption with minimal input perturbations.
MBZUAI ·
Researchers at National Taiwan University are developing low-complexity neural network technologies using quantization to reduce model size while maintaining accuracy. Their work includes binary-weighted CNNs and transformers, along with a neural architecture search scheme (TPC-NAS) applied to image recognition, object detection, and NLP tasks. They have also built a PE-based CNN/transformer hardware accelerator in Xilinx FPGA SoC with a PyTorch-based software framework. Why it matters: This research provides practical methods for deploying efficient deep learning models on resource-constrained hardware, potentially enabling broader adoption of AI in embedded systems and edge devices.